Fig. 5: Correction of phase errors from clock 1.
From: Dynamical decoupling of laser phase noise in compound atomic clocks
Phase errors from clock 1 are corrected using the measurement of clock 2. The resulting distribution of residual phase errors in the compound clock is estimated from that observed in clock 1: a Excitation probability pe in clock 2 as a function of the phase error Δϕ1 introduced by clock 1. b Probability distribution of phase errors Δϕ1 from clock 1, using the model discussed in the main text. c Probability distribution of residual phase errors Δϕtot of the compound clock, i.e. after correction by clock 2. Colour-coding illustrates which intervals of phase errors in b and c correspond to which fringe in a. Grey lines spanning all panels illustrate how phase errors from clock 1 are corrected by clock 2 and how the probability distribution in c is derived from the respective distribution shown in b. Vertical arrows in c indicate Dirac delta functions. Their integrals correspond to those of the respective shaded areas in b. As shown here, phase errors from clock 1 are corrected fully by clock 2 in nearly all cases. Additional noise introduced by clock 2 is treated separately (see main text).
